Abstract:

Objective

To identify the predictors of clinical efficacy of the tumour necrosis factoralpha （TNF-α） inhibitor adalimumab in the treatment of patients with rheumatoid arthritis （RA）.

Methods

A retrospective study was conducted on 216 RA patients treated with adalimumab who visited Jiangsu Province（Suqian） Hospital from January 2018 to December 2023， including 148 with response and 68 with no response to the therapy. Correlation analysis of the effectiveness of adalimumab in the treatment of RA was carried out using the point biserial correlation test and Spearman's Rho test. The predictors of efficacy of adalimumab for RA were identified using binary logistic regression analysis， and the predictive value of the predictors was assessed by receiver operating characteristic （ROC） curve analysis.

Results

The proportion of women（χ²=48.430， P＜0.001）， DAS28 （t=6.827， P＜0.001）， tumour necrosis factor-alpha （TNF-α） （t=3.719，P＜0.001）， and interferon-alpha （IFN-α） （t=5.22， P＜0.001） were higher in the response group than in the non-response group， while the non-response group had a higher proportion of patients with osteoporosis（χ²=56.263， P＜0.001） and glucocorticoid use （χ²=93.122， P＜0.001）. DAS28 （r=0.423， P＜0.001），TNF-α （r=0.171， P=0.012）， and IFN-α （r=0.237， P＜0.001） were positively associated with clinical effectiveness， whereas men （r=-0.427， P＜0.001）， osteoporosis （r=-0.510， P＜0.001）， and glucocorticoid therapy（r=-0.657， P＜0.001） were negatively associated with clinical effectiveness. The results of regression analyses showed that being female ［OR （95%CI）=0.001 （0， 0.067， P＜0.001］， no osteoporosis ［OR （95%CI）=0（0， 0.049）， P=0.003］， high DAS28 ［OR （95%CI）=83.455 （5.225， 1332.91）， P=0.002］， high IFN-α ［OR（95%CI） =10.259 （2.28， 46.164）， P=0.002］， and no glucocorticoid treatment ［OR （95%CI）=0.019 （0.001，0.366） P=0.019］ may be predictive factors for patients with RA responding to adalimumab treatment. With a cutoff value of 3.665， DAS28 had an area under the curve （AUC） value of 0.724 for the prediction of treatment response， with a sensitivity of 44.12% and specificity of 98.65%. With a cutoff value of 3.750，IFN-α had an AUC value of 0.780 for the prediction of treatment response， with a 97.06% sensitivity and 57.82% specificity. The AUC for the combined DAS28 and IFN-α was 0.898， and the sensitivity and specificity were 93.75% and 73.44%， respectively.

Conclusion

Female gender， no osteoporosis， high DAS28， high IFN-α， and no glucocorticoid treatment may be predictors of RA patients responding to adalimumab treatment. Among them， DAS28 is a more specific predictor， IFN-α is a more sensitive predictor，and the combination of the two has greater predictive value.

Key words: TNF-α antagonist, Adalimumab, Rheumatoid arthritis, Clinical efficacy, Predictive factors